I have a friend who has been using fMRI for more than 10 years and now a professor. He heard NIRS might be a good technology to measure human brain but knew little about it. So what is the difference between NIRS and fMRI? This is probably a common question for fMRI researchers who are looking for alternative ways to do brain imaging.
So here is some side-by-side comparison between NIRS and fMRI:
 NIRS |
 fMRI |
Note | |
|---|---|---|---|
| Measurement | both oxy- and deoxy-Hb |
BOLD signal |
That means you get two number per location at a time in NIRS |
| Temporal resolution | 10Hz | 0.5Hz | In fMRI TR is usually 2s |
| Spatial resolution | 30mm | 2-4mm | |
| Measurable region | Only surface |
Entire brain |
|
| Real-time application | easy | difficult | |
| Real-life application | easy | difficult | such as exercise, natural conversation etc |
| Hyperscanning application | easy | difficult | |
| Cost | Cheaper $100-500K |
Expensive >$2M |
NIRS price largely depends on the number of channels |
| Maintenance Cost | Cheap | Expensive | NIRS has ~0 maintenance cost; fMRI usually requires a on-site expert |
| Portability | Can be |
No | There are some portable NIRS devices available |
| Signal to noise ratio | lower | higher | |
| Subjects with Dental Braces and Retainers, or ear rings | Safe | might not be safe | |
| Subjects with Pacemaker or metals in brain | Safe | unsafe |
